✉news TechnologyInternet first seen 1 d ago, last 11 h ago, peak #2
Anthropic Pulls Its AI Model Evals Offline
Original: Anthropic Is Banishing Its Model Evals From the Internet
AI company Anthropic is removing its model evaluations from the internet, according to a report by Gizmodo. The move means the benchmarks and tests the company uses to measure its AI models' safety and performance will no longer be publicly accessible, limiting outside scrutiny of how the systems are assessed. Commenters in tech circles are weighing what this means for transparency in an industry already criticized for closing off safety research.
Why now: Reduced transparency from a leading AI lab raises concerns about independent oversight of frontier AI safety testing.
Rank over time, top of the chart is #1. 17 snapshots from 1 d ago to 11 h ago.
Evidence
API: https://socialmediatrends-api.osmike.com/v1/trends/1792328